Usage-based Traffic Control for P2P Content Delivery
Commenced in January 2007
Frequency: Monthly
Edition: International
Paper Count: 33090
Usage-based Traffic Control for P2P Content Delivery

Authors: Megumi Shibuya, Tomohiko Ogishi

Abstract:

Recently, content delivery services have grown rapidly over the Internet. For ASPs (Application Service Provider) providing content delivery services, P2P architecture is beneficial to reduce outgoing traffic from content servers. On the other hand, ISPs are suffering from the increase in P2P traffic. The P2P traffic is unnecessarily redundant because the same content or the same fractions of content are transferred through an inter-ISP link several times. Subscriber ISPs have to pay a transit fee to upstream ISPs based on the volume of inter-ISP traffic. In order to solve such problems, several works have been done for the purpose of P2P traffic reduction. However, these existing works cannot control the traffic volume of a certain link. In order to solve such an ISP-s operational requirement, we propose a method to control traffic volume for a link within a preconfigured upper bound value. We evaluated that the proposed method works well by conducting a simulation on a 1,000-user scale. We confirm that the traffic volume could be controlled at a lower level than the upper bound for all evaluated conditions. Moreover, our method could control the traffic volume at 98.95% link usage against the target value.

Keywords: P2P, traffic control, traffic localization, ALTO.

Digital Object Identifier (DOI): doi.org/10.5281/zenodo.1329707

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1562

References:


[1] Cisco Systems Inc., "Cisco Visual Networking Index: Forecast and Methodology, 2009-2014", http://www.cisco.com/en/US/solutions/collateral/ns341/ns525/ns537/ns7 05/ns827/white_paper_c11-481360_ns827_Networking_Solutions_Whit e_Paper.html.
[2] BitTorrent, Inc., BitTorrent DNA, http://www.bittorrent.com/dna.
[3] Pando Web Page, http://www.pando.com/.
[4] A. Nakao, K. Sasaki, and S. Yamamoto, "A Remedy for Network Operators against Increasing P2P Traffic: Enabling Packet Cache for P2P Applications", IEICE Trans. on Communications, pp.3810-3820, 2008.
[5] C. Hsu, and M. Hefeeda, "ISP-Friendly Peer Matching without ISP Collaboration", Proc. Int. Workshop on ROADS'08, December 2008.
[6] H. Xie, Y. R. Yang, A. Krishnamurthy, Y. Liu, and A. Silberschatz, "P4P: Provider Portal for Applications", Proc. ACM SIGCOMM-08, pp.351-362, August 2008.
[7] M. Shibuya, T. Ogishi, and S. Yamamoto, "A Study on Traffic Localization Issue for P2P Content Delivery", Proc. IMSA 2009, pp.26-32, August 2009.
[8] MaxMind, Inc., GeoIP Web Page, http://www.maxmind.com/app/ip-location.
[9] http://www.bittorrent.org/beps/bep_0003.html.
[10] M. Shibuya, T. Ogishi, and T. Kato, "Evaluation of Content Delivery Characteristics using Network-aware P2P Simulator", IEICE technical Report, IN2009-155, pp.67-72, March 2010. (in Japanese)
[11] IETF ALTO WG, http://tools.ietf.org/wg/alto/.